Proc. Natl. Acad. Sci. USA Vol. 93, pp. 3984-3989, April 1996 Neurobiology The ALIAmide and , but not , are protective in a delayed postglutamate paradigm of excitotoxic death in cerebellar granule neurons (N-methyl-D-aspartate/neurotoxicity/N-acylethanolamides/neuroprotection/) S. D. SKAPER*, A. BURIANI, R. DAL Toso, L. PETRELLI, S. ROMANELLO, L. FACCI, AND A. LEON Researchlife S.c.p.A., Centro di Ricerca Biomedica-Ospedale Civile, 31033 Castelfranco Veneto (TV), Italy Communicated by Rita Levi-Montalcini, Consiglio Nazionale Richerche, Rome, Italy, January 2, 1996 (received for review December 16, 1995) ABSTRACT The amino acid L-glutamate is a neurotrans- ceptors may be responsible for much of the neuronal damage mitter that mediates fast neuronal excitation in a majority of associated with certain acute insults, including hypoxia- synapses in the . Glutamate stimulates ischemia, hypoglycemia, epilepsy, and trauma (5, 6). Further- both N-methyl-D-aspartate (NMDA) and non-NMDA recep- more, exaggerated EAA receptor activity has been suggested tors. While activation of NMDA receptors has been implicated by some to also underly chronic neurodegenerative disorders, in a variety of neurophysiologic processes, excessive NMDA including Huntington disease, Alzheimer disease, Parkinson receptor stimulation (excitotoxicity) is thought to be primarily disease, amyotrophic lateral sclerosis, and acquired immune responsible for neuronal injury in a wide variety of acute deficiency syndrome-dementia complex (7-9). In many ex- neurological disorders including hypoxia-ischemia, seizures, perimental systems, overstimulation of the N-methyl-D- and trauma. Very little is known about endogenous molecules aspartate (NMDA) subtype of glutamate receptor with pro- and mechanisms capable of modulating excitotoxic neuronal tracted entry of Ca2+ into neurons appears to be a principal death. Saturated N-acylethanolamides like palmitoylethanol- mechanism for subsequent damage (5). There is growing amide accumulate in ischemic tissues and are synthesized by evidence for participation of non-NMDA receptors in EAA- neurons upon excitatory amino acid receptor activation. Here mediated neurotoxicity as well, especially in cases of prolonged we report that palmitoylethanolamide, but not the cognate or chronic insult (10, 11). N-acylamide anandamide (the ethanolamide of arachidonic N-acylethanolamides, like palmitoylethanolamide, and N- acid), protects cultured mouse cerebellar granule cells against acylphosphatidylethanolamides accumulate in conditions in- glutamate toxicity in a delayed postagonist paradigm. Palmi- volving degenerative changes to tissues (12), including brain toylethanolamide reduced this injury in a concentration- (13) and cardiac (14) ischemia. Furthermore, EAAs can dependent manner and was maximally effective when added stimulate the synthesis of N-acylethanolamides and N- 15-min postglutamate. Cannabinoids, which like palmi- acylphosphatidylethanolamides in cultured central neurons toylethanolamide are functionally active at the peripheral (15, 16). The possibility that compounds of this type could receptor CB2 on mast cells, also prevented defend against an excitotoxic insult may be entertained (16). neuron loss in this delayed postglutamate model. Further- Interestingly, it has recently been reported that mast cells, more, the neuroprotective effects of palmitoylethanolamide, multifunctional immune cells implicated in immediate hyper- as well as that of the active cannabinoids, were efficiently sensitivity and inflammatory reactions (17), express a periph- antagonized by the candidate central eral-type cannabinoid receptor (designated CB2; ref. 18) that (CB1) agonist anandamide. Analogous pharmacological be- recognizes palmitoylethanolamide (ALIAmides) and down- haviors have been observed for palmitoylethanolamide (ALI- modulates activation of these cells in vitro (19). The candidate Amides) in downmodulating activation. Cerebellar endogenous agonist for the brain cannabinoid receptor CB1 granule cells expressed mRNA for CB1 and CB2 by in situ (20, 21), arachidonylethanolamide (anandamide) (22, 23), hybridization, while two cannabinoid binding sites were de- binds to mast cell CB2 and actually antagonizes the functional tected in cerebellar membranes. The results suggest that (i) effects of palmitoylethanolamide and several cannabinoids (19). non-CB1 cannabinoid receptors control, upon agonist binding, We now report that palmitoylethanolamide and 2-O-(83-D- the downstream consequences of an excitotoxic stimulus; (ii) glucopyranosyl)-N-palmitoylethanolamide, as well as some palmitoylethanolamide, unlike anandamide, behaves as an en- natural and but not anandamide, are dogenous agonist for CB2-like receptors on granule cells; and efficacious in protecting cultured cerebellar granule cells from (iii) activation of such receptors may serve to downmodulate glutamate toxicity in a delayed postagonist paradigm without deleterious cellular processes following pathological events or affecting EAA receptor function. Anandamide, however, an- noxious stimuli in both the nervous and immune systems. tagonized these neuroprotective effects. Furthermore, granule cells expressed mRNA for both CB1 and CB2 and cerebellar membranes two cannabinoid sites. Dicarboxylic amino acids form the most widespread excitatory displayed binding transmitter network in the mammalian brain (1). Glutamate interactions with specific membrane receptors are responsible MATERIALS AND METHODS for many neurologic functions, including cognition, memory, Cultures. Cultures In the excitation Primary Neuron containing granule movement, and sensation (2). addition, neurons were prepared from dissociated cerebella of 7- to produced by glutamate is important in influencing the devel- 8-day-old BALB/c mice (Modelli Biologici Sperimentali, Tre- opmental plasticity of synaptic connections in the nervous Cells were in basal medium system (3, 4). Excitatory amino acids (EAAs) have also been viso, Italy) (24). plated Eagle's implicated in neurotoxicity. Excessive activation of EAA re- Abbreviations: EAA, excitatory amino acid; NMDA, N-methyl-D- aspartate; DIV, days in vitro; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5- The publication costs of this article were defrayed in part by page charge diphenyltetrazolium bromide; PMSF, phenylmethylsulfonyl fluoride; payment. This article must therefore be hereby marked "advertisement" in KA, kainic acid. accordance with 18 U.S.C. §1734 solely to indicate this fact. *To whom reprint requests should be addressed. 3984 Downloaded by guest on September 30, 2021 Neurobiology: Skaper et al. Proc. Natl. Acad. Sci. USA 93 (1996) 3985 supplemented with 10% fetal calf serum (BIOSPA, Wedel, weeks at 4°C, developed with Phenisol (Ilford), fixed with Germany), 25 mM KCl, 2 mM glutamine, and gentamicin (50 Hypam (Ilford), and counterstained with cresyl violet. tag/ml) on 35-mm-diameter dishes (Falcon) coated with poly- Radioligand Binding Assays. Cerebella from 20-day-old (L-lysine) (10 Atg/ml) (Mr, 68,000), 2.5 x 106 cells per dish. BALB/c mice were removed, cleaned of meninges, and stored Cytosine P3-D-arabinofuranoside (10 JLM) was added to the at -80°C for up to 1 month. Membrane preparation and culture medium 18-20 h after plating to halt nonneuron binding assays were carried out following published proce- growth. The cultures were used at 8-10 days in vitro (DIV), and dures (19, 31), with modification. Groups of five (frozen) contain -95% glutamatergic granule neurons (25). cerebella were homogenized in 20 ml of binding buffer (3 mM Induction of Glutamate Neurotoxicity. Neurotoxicity was MgCl2/1 mM EDTA/50 mM Tris, pH 7.4) containing 0.32 M induced essentially as described (24). Medium from 8- to sucrose. The homogenate was centrifuged at 2000 x g for 10 10-day-old granule cell cultures was removed and saved. min at 4°C, and the resulting supernatant was centrifuged at Culture dishes were washed twice with Mg2+-free Locke's 15,000 x g for 15 min at 4°C. The pellet (P2) was resuspended solution, and the cells were then incubated with 500 ,iM in 20 ml of binding buffer containing 0.1% fatty acid-free glutamate in Mg2+-free Locke's solution for 5 min (23-25°C). bovine serum albumin, and the last centrifugation step was The glutamate-containing solution was then removed by as- repeated. The final pellet was gently rinsed with distilled water piration, and the dishes were washed twice with complete and then resuspended in 1 ml ofbinding buffer (0.8-1.1 mg per Locke's solution and then returned to the incubator in their ml of protein). Membranes were kept on ice and used within original medium for a further 24 h. Acute glutamate exposure 1 h. Binding experiments were performed in silicon-treated under these conditions consistently resulted in a 50-65% loss tubes. [3H]WIN 55,212-2 (45.5 Ci/mmol; 22 ttM in ethanol; in neuron numbers 24 h later, at which time cell survival was New England Nuclear) and nonradioactive ligands were seri- routinely evaluated. Drug treatment protocols are described in ally diluted in dimethyl sulfoxide and were added at the desired the appropriate figure or table legend as introduced. Stock concentration to a final volume of 500 ,ul ofbinding buffer. The solutions ofcannabinoids and N-acylethanolamides were made final concentration of dimethyl sulfoxide was always 1%. in dimethyl sulfoxide, while anandamide was dissolved in Binding was initiated by adding 30 ,ug (protein) of membranes ethanol. The final concentration of solvent in the culture never and the tubes were incubated for 55 min at 30°C. Binding was exceeded 0.2%, except in the case of palmitoylethanolamide, terminated by transferring the reaction mixture to an Eppen- which necessitated a higher concentration (1%). These con- dorf tube, followed by addition of fatty acid-free bovine serum centrations of solvent were found to have no effect on the albumin to 0.1% and centrifugation at 39,000 x g for 10 min at response of the granule cells to glutamate. 20°C. The supernatants were collected and counted to determine Quantitation ofNeurotoxicity. Glutamate neurotoxicity was the concentration offree , while the pellets were suspended gross enough to be evident morphologically when viewed in 1% Triton X-100/ethanol (1:1; vol/vol) and radioactivity was under a phase-contrast microscope. Neuronal survival was assayed by liquid scintillation counting. Nonradioactive WIN quantified using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltet- 55,212-2 (1 ,tM) was used to displace specific binding. razolium bromide (MTT), which yields a blue formazan prod- Materials. Tissue culture media and supplements and can- uct in living cells but not in dead cells or their lytic debris (26). nabidiol were obtained from Sigma. WIN 55,212-2 and A8- The reaction product, solubilized in dimethyl sulfoxide, is THC were from Research Biochemicals (Natick, MA). Palmi- easily measured with an ELISA plate reader and is directly toylethanolamide, 2-0-(f3-D-glucopyranosyl)-N-palmitoyleth- proportional to the number of neurons present (24, 27, 28). anolamide (glucosylpalmitoylethanolamide), and anandamide The MTT technique is equivalent to lactate dehydrogenase were synthesized by standard chemical techniques, with purity release in the measurement of excitotoxin-mediated neuronal of >99.5% as assessed by HPLC. All other reagents, unless death in vitro (29). specified otherwise, were from Sigma. In Situ Hybridization. BALB/c mouse (4-week-old) brain was snap-frozen in 2-methylbutane, 12-,im coronal sections RESULTS cut with a Jung model CM 3000 cryostat and thaw-mounted on polylysine-coated glass slides. All sections were then fixed in Palmitoylethanolamide- and Cannabinoid-Mediated Pro- 4% paraformaldehyde, dehydrated in a graded series of eth- tection from Glutamate Neurotoxicity. Although glutamate is anols, incubated 5 min in chloroform, and air-dried. Cerebellar a neuroexcitant agonist at NMDA, quisqualate, and kainate granule cells growing on polylysine-coated cover glasses at 8 postsynaptic receptors (32), its neurotoxicity is predominantly DIV were fixed with 4% paraformaldehyde, washed twice with mediated by NMDA receptors alone (33). Given that palmi- phosphate-buffered saline, and permeabilized in 70% ethanol toylethanolamide accumulates in ischemic brain (13) and its for 48 h at 4°C. The cells were then dehydrated using a higher production is induced in central nervous system neurons by graded series of ethanols and air-dried. Detection of CB1- and EAA receptor activation (15, 16), it was of interest to examine CB2-specific mRNA made use of the following oligonucleo- possible modulatory effects of the N-acylethanolamide against tides prepared with a Beckman model Oligo 1000 DNA NMDA receptor-mediated neurotoxicity. Most studies on the synthesizer: CB1, 5'-GGT GAC GAT CCT CTT ATA GGC neuroprotective efficacy of glutamate antagonists or other CAG AGG CCT TGT AAT GGA TAT GTA-3' (21); CB2, drugs have used pre- and cotreatment protocols. The present 5'-GGT GAC GAG AGC TTT GTA GGT AGG TGG GTA experimental setting was chosen, as others have done (34, 35), GCA CAG ACA TAG GTA-3' (18). A random sequence was to allow for conditions that provide potentially rescuable used as control. Oligonucleotides were 3' end-labeled with neurons in a delayed postagonist paradigm. dATP[a-35S] (1136 Ci/mmol; 1 Ci = 37 GBq; New England Palmitoylethanolamide (100 ,tM) or MK-801 (10 ,tM) was Nuclear) by terminal deoxynucleotidyltransferase (Pharma- added during or at various times after conclusion of a 5-min cia) to a specific activity of 109 cpm/,tg. All cover glasses glutamate exposure. Neuronal cell loss was quantitatively were hybridized in standard solutions (30) with 1.5 X 107 monitored by the MTT reaction, which is widely used to assess dpm/ml overnight in a humidified chamber at 42°C. The cover neuron numbers in culture (24, 27, 28) and is considered glasses were then washed once with lx SSC/0.1% SDS (30 equivalent to lactate dehydrogenase release (29). Microscopic min, 55°C), twice with lx SSC (15 min, 55°C), and once with observation of the cultures confirmed the results obtained by 0.1x SSC (30 min, 25°C), followed by a 2-min rinse in MTT (data not shown). The potent noncompetitive NMDA autoclaved water and dehydration with a graded series of antagonist MK-801 (36) while maximally efficacious when ethanols. The air-dried cover glasses were then dipped in K.5 present during glutamate exposure rapidly decayed with a photoemulsion (Ilford) (diluted 1:1 with water), exposed for 5 15-min postagonist delay (Fig. 1) (see also ref. 34). In contrast, Downloaded by guest on September 30, 2021 3986 Neurobiology: Skaper et al. Proc. Natl. Acad. Sci. USA 93 (1996) palmitoylethanolamide achieved maximal efficacy when added Table 1. Palmitoylethanolamide and cannabinoids 15 min postglutamate (Fig. 1). Neuroprotection produced by concentration-dependently rescue cerebellar granule cells from the delayed, postagonist addition of palmitoylethanolamide glutamate toxicity with late addition: Antagonism by anandamide was also concentration dependent (Table 1) and related to the EC50, ,LM duration of drug exposure, being maximal when present for 20 min or more. Addition of a glucose residue to palmitoyleth- Compound Control + Anandamide anolamide (glucosylpalmitoylethanolamide) increased its ac- Palmitoylethanolamide 54.6 + 15.3 >150 tivity as a neuroprotectant (Table 1), perhaps due to improved Glucosyl-PEA 12.1 ± 1.8 33.4 ± 3.5* solubility. The potent non-NMDA receptor antagonist CNQX 3.9 ± 1.1 15.5 ± 0.6* (11), Ca2+ channel blockers (diltiazem, nifedipine, o-cono- A8-THC 2.8 ± 0.7 16.9 ± 5.1* toxin, diphenylhydantoin, flunarizine), free radical scavengers, 11OH-A9-THC 0.88 ± 0.28 ND and monosialogangliosides were also tested in this delayed WIN 55,212-2 24.5 ± 4.1 ND postagonist rescue paradigm and found to produce little or no > 100 ND reduction in neuronal injury (data not shown). Anandamide > 100 ND Given that cannabimimetics display functional effects anal- Cannabinoids were added to the cultures for a 10-min period and ogous to palmitoylethanolamide on mast cells (19), cannabi- palmitoylethanolamide or glucosylpalmitoylethanolamide (glucosyl- noid compounds were also explored for possible protection PEA) was added for a 30-min period, all starting 15 min postglutamate. against glutamate neurotoxicity. The synthetic cannabinoid All incubations contained 150 ,iM PMSF to inhibit anandamide nabilone (10 tLM) and A8-THC (10 taM), like saturated N-acyl- metabolism. Neuron survival was assessed 24 h later. EC5o is the were effective with a 15- concentration reducing by 50% the cell death caused by glutamate. ethanolamides, maximally delayed, Values are means ± SD from at least three experiments. Anandamide min postagonist application (Fig. 1); exposure times of 5-10 was used at 10 ,LM. ND, not determined. min proved to be optimally efficacious. The respective EC50 *P < 0.01 vs. control. values are given in Table 1. In contrast, the nonpsychoactive cannabinoid cannabidiol, which has weak affinity for CB1 and ide congener homo-,y-linolenylethanolamide (C20:3), which CB2 receptors (18, 20), was inactive (Table 1). binds the CB1 receptor more weakly than anandamide (23), Anandamide Antagonizes the Neuroprotective Effects of antagonized neither CB2 receptor function in mast cells nor Saturated N-Acylethanolamides and Cannabinoids. Ananda- the rescue effects of saturated N-acylethanolamides and nab- mide, a candidate agonist for the brain cannabinoid receptor ilone against glutamate neurotoxicity when used up to 25 juM (22, 23), produces many of the behavioral and physiological (data not shown), indicating a degree of specificity in the responses of cannabinoids attributed to activation of the CB1 observed behaviors of anandamide. receptor (37, 38). Unlike the active cannabinoids tested (Table When granule cell incubation with anandamide (50 ,tM) was 1), a 10-min anandamide exposure of up to 100 ,tM (in the prolonged beyond 1 h in the presence of 150 ,iM PMSF, an presence of 150 ,M phenylmethylsulfonyl fluoride; PMSF) inhibitor of anandamide amidase (39), neuron survival was failed to protect granule cells from glutamate toxicity. When reduced to 30% ± 12% (n = 6) of control 24 h later. The added together with saturated N-acylethanolamides or canna- potent cannabimimetic compound HU-210 reportedly is cy- binoids, anandamide (10 ,tM) actually antagonized their de- totoxic for neurons in vitro, an effect attributed to its interac- layed postagonist neuroprotective effects, shifting the concen- tion with the CB1 receptor (40). Active cannabinoids, but not tration-response curve for the drug (Table 1). The anandam- cannabidiol, were also cytotoxic to granule cells when exposure to high (-25 piM) concentrations exceeded 60 min: 50 ,M 120- nabilone or A8-THC decreased survival by 80-85% after 24 h. Incubation of granule neurons with 100 ,uM palmitoylethanol- amide or its glucosyl derivative for 24 h did not affect cell vitality. 100 - Palmitoylethanolamide and Cannabinoids Do Not Antago- nize EAA Receptor Function in Intact Neurons. The delayed 80- postglutamate neuroprotection afforded by saturated N- acylethanolamides and cannabinoids (Table 1) made unlikely a direct effect at the EAA receptor level, as confirmed here. 60 The delayed component of kainic acid (KA)-mediated neu- ronal death involving free radical production especially evi- 40 dent under conditions that eliminate the occurrence of acute ion-dependent neuronal swelling and cell lysis (40) was inves- tigated. When mature granule cells were treated with 500 utM 20 KA for 30 min in a Na+-free buffer, the 75% loss seen 2 h later was, as expected (41), largely prevented by CNQX or the 0- v- antioxidant butylated hydroxytoluene (Table 2). In contrast, CTRL - 5 0 1'5 30 neither palmitoylethanolamide nor cannabinoids were neuro- protective when present during and after KA exposure in these MIN AFTER GLUTAMATE EXPOSURE conditions (Table 2). Palmitoylethanolamide and CNQX, but FIG. 1. Time course of rescue from glutamate neurotoxicity: not butylated hydroxytoluene, were cytoprotective for granule palmitoylethanolamide and cannabinoids vs. MK-801. Sister cultures cells incubated with 500 ,tM KA in Na+-containing buffer, where were exposed to 500 ,iM glutamate for 5 min and then rescued by rapid lysis of =50% of the neurons occurs during the 30-min drug addition of 10 ,iM MK-801 (0), 10 ,uM nabilone (A), 10 ,uM A8-THC exposure period (42). Thus, a KA receptor antagonist (CNQX) (D), or 100 ,iM palmitoylethanolamide (A) at the indicated time in was able to reduce both components of KA receptor-triggered minutes after washout of the glutamate. 0, Glutamate alone. PMSF neuronal injury while the N-acylethanolamide affected only the (150 ,uM) was included in all incubations. Except for those cultures in excitotoxic that the latter which the was on branch, indicating operates indepen- drug present only during glutamate exposure (-5 the of the EAA time axis), the duration of late addition drug treatment was as follows: dently receptor. MK-801, 30 min; palmitoylethanolamide, 30 min; cannabinoids, 10 Cerebellar granule cell development in vitro is strictly de- min. Neuronal survival by MTT was measured 24 h later. Values are pendent on Ca2+ influx, mediated by either voltage- or NMDA means ± SD (three experiments). receptor-gated channels (43, 44). Predictably, MK-801 negated Downloaded by guest on September 30, 2021 Neurobiology: Skaper et al. Proc. Natl. Acad. Sci. USA 93 (1996) 3987

Table 2. Palmitoylethanolamide and cannabinoids do not interfere A * t with KA receptor activity in cerebellar granule cells in the absence of Na+ Culture conditions % survival *_,__ v._F._- _ KA 25.6 ± 3.1 ^ ^ s X s _ v il_iP KA+CNQX 87.3± 11.1 * j_SS.fi;_ ';iiliP--,:-r . _! KA + BHT 84.6 ± 10.0 _s,.s,., KA + PEA 35.0 ± 4.2 __-g!=.,:t,.. .: o_a_...... * KA + nabilone 36.2 ± 4.9 Nine-day-old cultures were incubated for 30 min (27°C) in Na+-free Locke's solution containing 500 ,uM KA and one of the following: 10 B ._. 4. ... s, ,uM CNQX, 100 ,uM butylated hydroxytoluene (BHT), 100 ,uM a . palmitoylethanolamide (PEA), 10 ,uM nabilone, or 3 ,uM 11OH-A9- *...... : THC. Cultures were then washed free of KA and incubated 2 h more 4:* :: (27°C) in Na+-free Locke's solution with the indicated compound. PMSF (150 ,uM) was present during all drug incubations. Cell survival was quantified at the end of the 2 h. Values are means ± SD (three experiments). the trophic effect of NMDA for granule cells cultured 7 days in medium containing a reduced concentration (5 mM) of KCI. IC _1 In contrast, the presence of either glucosylpalmitoylethanol- I) amide or 11OH-A9-THC throughout the same 7-day period (in place of MK-801) did not interfere with the NMDA trophic action (Table 3). The latter cannabimimetic was chosen be- cause of its minimal long-term cytotoxicity. Palmitoylethanol- amide was not tested here, because ofthe deleterious influence i...... of long-term cell exposure to high solvent concentrations. Cerebellar Granule Cells and Cerebellum Express the I Genes Encoding Cannabinoid Receptors CB1 and CB2. The OP. observed effects of cannabinoids and palmitoylethanolamide FIG. 2. In situ hybridization of cannabinoid receptor mRNA in in downmodulating EAA neurotoxicity, and their antagonism cultured mouse cerebellar granule cells (A-C) and 4-week cerebellum by anandamide, prompted the question of whether cerebellar (A'-C'). (A and A') CB1. (B and B') CB2. (C and C') Random. (Bar granule cells express cannabinoid receptor gene transcripts. In - 10 Jim.) situ hybridization of 8 DIV granule cell monolayers using CB1- and CB2-specific oligonucleotide probes revealed the presence of [3H]WIN 55,212-2 to membranes from mouse cerebellum, of mRNA for both receptor types in neuronal perikarya (Fig. with two apparent binding sites. The Scatchard (Rosenthal) 2 A-C). The CB1 and CB2 mRNAs were detectable in the plot is shown in Fig. 3. Values for Kdl of 1.6 ± 1.0 nM and Kd2 majority of granule cells examined, suggesting the same neu- of 11.0 ± 1.5 nM, with corresponding Bmx of 0.7 ± 0.1 and 3.2 ron to be capable of simultaneously expressing the two known ± 0.3 pmol per mg of protein, respectively, were obtained by cannabinoid receptor subtypes. It was not possible, however, to analyzing the data according to Rosenthal (45) and using the ascertain the relative abundance of the two mRNAs. Sections FIG. P computer program for a two binding site fitting (BIO- of mouse cerebellum, when hybridized with CB1- and CB2- SOFT, Cambridge, U.K.). Prior studies of ligand binding to specific probes, also revealed the presence of the correspond- cannabinoid receptors in adult rat brain have described a single ing mRNAs in the granule cell layer (Fig. 2 A'-C'). The Purkinje cell layer, while producing a positive hybridization signal for CB2, was negative for CB1, indicating the ability of 90- this in situ procedure to discriminate between different subsets 7 5- of neurons in the same tissue section. 0.06 60- Cerebellum Contains Two Cannabinoid Binding Sites. Ra- lm45.- z demonstrated 0.05 o 30 dioligand binding experiments specific binding 0 Ca 15- Table 3. Glucosylpalmitoylethanolamide and cannabinoids do not 0.04 interfere with NMDA receptor function in developing cerebellar 0 3000 6000 9000 12000 1S000 18000 aII FREE pM granule cells . , 0.03 Culture conditions % survival

K25 100 0.02 . KS plus U None 31.6 ± 3.8 _ _ 0.01 - U NMDA 69.6 ± 6.1 NMDA + MK-801 33.3 ± 4.2 NMDA + glucosyl-PEA 69.5 ± 1.5 0 25 50 75 100 125 150 175 200 NMDA + 11OH-A9-THC 70.3 ± 2.9 B Cultures were incubated from days 1-8 in medium containing 25 mM KCl (K25) or 5 mM KCl (KS) and the indicated additions: NMDA, FIG. 3. Scatchard plot of specific [3H]WIN 55,212-2 binding to 150 ,uM; MK-801, 10 ,uM; glucosylpalmitoylethanolamide (glucosyl- cerebellar membranes. Specific binding was defined as the difference PEA), 50 ,tM; 11OH-A9-THC, 3 ,uM. The last two compounds were between binding that occurred in the presence and absence of 1 JIM resupplied daily. Cell survival was quantified at the end of the eighth nonradioactive ligand. (Inset) Saturation isotherm. Data are means ± day. Values are means ± SD (three experiments). SD (seven experiments). B, bound (pM); F, free. Downloaded by guest on September 30, 2021 3988 Neurobiology: Skaper et al. Proc. Natl. Acad. Sci. USA 93 (1996) binding site (31, 44). In these latter experiments, however, agonist rescue effects from a CB1-mediated process. It is tempt- steps were taken (preincubation and washing) that were re- ing to speculate that saturated (ALIAmides) and unsaturated ported necessary for observing a homogeneous binding site (anandamides) long-chain fatty acid ethanolamides may be can- (31), or that may have masked the presence of a higher affinity nabinoid receptor type-specific endogenous agonists. site by the use of albumin (46). In initial trials anandamide fully The ability of glucosylpalmitoylethanolamide to prevent displaced bound radioligand, while palmitoylethanolamide excitotoxic neuronal injury in a delayed postagonist setting, seemed to be only partially effective (data not shown). Given and with greater efficacy than the parent molecule, may be that WIN 55,212-2 has similar affinities for CB1 and CB2 (18, attributable to one or several possibilities. Although modified 47), it is difficult to assign a site to palmitoylethanolamide, solubility may be a factor, metabolism or transformation of the although it is reported to not bind CB1 (22, 23). glucosyl derivative to palmitoylethanolamide or a related N- acylamide needs to be considered. This avenue is currentlybeing DISCUSSION explored. While the glucosylacylamide exhibited a pharmacolog- ical profile not unlike that ofthe startingcompound, it is presently The experiments described here assign a functional correlate not possible to draw conclusions as towhether or not the molecule to the naturally occurring saturated N-acylamide, palmi- per se is recognized by cannabinoid receptors. toylethanolamide, in central neurons. We have demonstrated The observed neuroprotective effects of palmitoylethanol- that (i) palmitoylethanolamide and cannabinoids, but not the amide and cannabinoids probably did not result from inter- unsaturated N-acylamide anandamide, downmodulate in a ference with EAA receptor function. These compounds, in delayed postagonist manner the toxic consequences of EAA contrast to NMDA antagonists, were only modestly protective receptor activation in cultured cerebellar granule cells; (ii) when added concurrently with glutamate, but they became anandamide antagonizes the neuroprotection afforded by more efficacious with increasing delay of postagonist intro- palmitoylethanolamide and cannabinoids; (iii) granule cells duction. Saturated N-acylethanolamides and cannabinoids, express the genes for CB1 and CB2; and (iv) [3H]WIN 55,212-2 however, reduced neither neuronal injury caused by kainate binding to cerebellar membranes displays two different bind- receptor-induced oxidative stress nor neuronal survival pro- ing affinities, suggesting the presence of at least two binding moted by NMDA receptor stimulation in immature cerebellar sites. Because palmitoylethanolamide, but not anandamide, granule cells. As expected, the latter two processes were appears to behave as an endogenous agonist for the CB2 sensitive to kainate and NMDA receptor-specific antagonists, receptor on mast cells and downregulates their activation in respectively. Moreover, palmitoylethanolamide and cannabi- vitro (19), the present findings suggest that a CB2-like receptor noids prevented neurotoxicity triggered by NMDA or by may also exert a negative regulatory effect on postglutamate kainate under conditions favoring excitotoxicity over free receptor events following excessive excitatory stimulation. radical toxicity. The nonpsychotropic cannabinoid HU-211 Palmitoylethanolamide, unlike anandamide, is reported to attenuated NMDA receptor-mediated toxicity to cultured not bind the CB1 receptor (22, 23). Anandamide, by binding cortical neurons, apparently by binding to NMDA receptors to brain CB1 (22, 23), produces many of the behavioral and (40). HU-211 was most effective when coapplied with EAA physiological responses ofcannabinoids (47). The psychotropic agonist (40). Antagonists of NMDA and kainate/AMPA effects of cannabinoids are presumably mediated via activation receptors, Ca2+ channel blockers, free radical scavengers, of the brain CB1 receptor (31), which is a typical member of inhibitors of protein and RNA synthesis, phosphatases and the G-protein-coupled superfamily of receptors (20, 21). Cul- synthase, and monosialogangliosides all failed to tured granule cells and cerebellum expressed mRNA for both show a delayed postagonist neuroprotective action (unpub- CB1 and CB2 by in situ hybridization but other receptor lished observations). The nonspecific endonuclease inhibitor variants could exist. The binding data with cerebellar mem- ATA attenuated glutamate neurotoxicity in cotreatment but branes are consistent with the presence of two binding sites not when added 15-min postglutamate, in difference to a with affinities close to those published for CB1 and CB2 (18, recent report of delayed ATA protection for cortical neurons 47). Although CB2 receptor identification until now has been (35). The N-acylamide and cannabinoid protective effects in limited to peripheral tissues (18, 19, 48), a detailed study of its the delayed postagonist paradigm required only a limited expression or that of other cannabinoid receptor forms in exposure window, suggesting these compounds to interfere different brain areas as a function of species/development is with one or more downstream consequences of excitotoxic lacking. G-protein-linked cannabinoid receptors have been glutamate receptor overaction. described to be present in rat cerebellar granule cells in vitro Given its clinical relevance, much effort has been directed to (49), although receptor subtypes were not evaluated. Neuronal devising pharmacological means of preventing or reducing the expression of CB2 could be regulated by culture conditions, or neuropathological consequences of exaggerated EAA recep- in vivo by injury, as in the case of some EAA receptor subtypes tor stimulation, mainly by antagonizing receptor activation, in vitro in cerebellar granule cells (50) and after cerebral blocking receptor- and voltage-gated ion channels, inhibiting ischemia (51). Palmitoylethanolamide, but not anandamide, EAA release, or by interfering with postreceptor processes was neuroprotective for granule cells; anandamide actually (53-55). An alternative strategy to mitigate excitotoxic neu- antagonized the protective action of the former. We recently ronal death would be to identify natural pathways whose reported that anandamide binds to the CB2 receptor on mast activation downregulates the deleterious outcome of toxic or cells and antagonizes the ability of palmitoylethanolamide and noxious stimuli. In this respect, it is of more than passing cannabinoids to inhibit mediator release (19). Anandamide interest that EAAs themselves can induce brain neurons in also inhibited specific [3H]WIN 55,212-2 binding to mast cell vitro to produce palmitoylethanolamide and other N- membranes (19), suggesting it to behave as a functional acylethanolamides (15, 16). Conceivably, certain of these antagonist at least for CB2 on mast cells. Such opposite molecules might play a role in modulating cellular defense behavior is typical of differences in the agonistic ability of mechanisms by acting at non-CB1 cannabinoid receptors, receptor ligands in cerebellar granule cells and, in analogy with much as they do in the case of mast cells (19). By providing the mast cells (19), suggests that the two N-acylethanolamides neuron with exogenous palmitoylethanolamide (or semisyn- have differing roles for different cannabinoid receptors. In- thetic saturated N-acylethanolamides), one might be making terestingly, cannabinoid receptor activation may inhibit the available quantities of its physiological modulator sufficient to presynaptic release of glutamate via an inhibitory G protein in restore cellular homeostasis in the face of an excitotoxic hippocampal neurons (52), suggesting a presynaptic CB1 lo- challenge. This pharmacologic approach has recently found cation, thus further distinguishing the present delayed post- application in the ability of palmitoylethanolamide to control Downloaded by guest on September 30, 2021 Neurobiology: Skaper et aL. Proc. Natl. Acad. Sci. USA 93 (1996) 3989 mast cell activation by a local autacoid antiinflammatory 25. Levi, G., Aloisi, F., Ciotti, M. T. & Gallo, V. (1984) Brain Res. mechanism, hence the term ALIA (19, 56, 57). Palmitoyleth- 290, 77-86. anolamide, in reality, appears to exert a more broad local 26. Mossmann, T. (1983) J. Immunol. Methods 65, 55-63. autacoid antiinjury function, thus the acronym autacoid local 27. Manthorpe, M., Fagnani, R., Skaper, S. D. & Varon, S. (1986) for ALIA. of this Dev. Brain Res. 25, 191-198. injury antagonism Therapeutic implications 28. Ohsawa, F., Widmer, H. R., Knusel, B., Denton, T. L. & Hefti, mechanism include the development of innovative neuropro- F. (1993) Neuroscience 57, 67-77. tective drugs for central nervous system injury. 29. Patel, J., Zinkand, W. C., Thompson, C., Keith, R. & Salama, A. (1990) J. Neurochem. 54, 849-854. We thank Dr. Gabriele Marcolongo for synthesis and purification of 30. Cosi, C., Spoerri, P. E., Comelli, C., Guidolin, D. & Skaper, S. D. the N-acylethanolamides, Mr. Michele Fabris for preparation of the (1993) NeuroReport 4, 527-530. graphs, Ms. Patrizia Lentola for secretarial assistance, and Modelli 31. Devane, W. A., Dysarz, F. A., III, Johnson, M. R., Melvin, L. S. Biologici Sperimentali for skilled technical assistance. We extend a & Howlett, A. C. (1988) Mol. Pharmacol. 34, 605-613. special note of thanks to the late Prof. Angelo Burlina, whose untiring 32. Watkins, J. C. & Olverman, H. J. (1987) Trends Neurosci. 10, efforts and initiative in biomedical research helped to make this work 265-272. possible. 33. Choi, D. W., Koh, J.-Y. & Peters, S. (1988) J. Neurosci. 8, 185-196. 1. Cotman, C.W., Monaghan, D.T., Offersen, O.P. & Storm- 34. Hartley, D. M. & Choi, D. W. (1989) J. Pharmacol. Exp. Ther. Mathisen, J. (1987) Trends Neurosci. 10, 273-280. 250, 752-758. 2. Gasic, G. P. & Hollmann, M. (1992) Annu. Rev. Physiol. 54, 35. Csernansky, C. A., Canzoniero, L. M. T., Sensi, S. L., Yu, S. P. & 507-536. Choi, D. W. (1994) J. Neurosci. Res. 38, 101-108. 3. Collingridge, G. & Bliss, T. V. P. (1987) Trends Neurosci. 10, 36. Wong, E. H., Kemp, J. A., Priestley, T., Knight, A. R., Woodruff, 288-294. G. N. & Iversen, L. L. (1986) Proc. Natl. Acad. Sci. USA 83, 4. Lipton, S. A. & Kater, S. B. (1989) Trends Neurosci. 12, 265-270. 7104-7108. 5. Choi, D. W. (1988) Neuron 1, 623-634. 37. Crawley, J. N., Corwin, R. L., Robinson, J. K., Felder, C. C., 6. Choi, D. W. & Rothman, S. M. (1990) Annu. Rev. Neurosci. 13, Devane, W. A. & Axelrod, J. (1993) Pharmacol. Biochem. Behav. 171-182. 46, 967-972. 7. DiFiglia, M. (1990) Trends Neurosci. 13, 286-289. 38. Smith, P.B., Compton, D.R., Welch, S.P., Razdan, R.K., 8. Meldrum, B. & Garthwaite, J. (1990) Trends Pharmacol. Sci. 11, Mechoulam, R. & Martin, B. R. (1994)J. Pharmacol. Exp. Ther. 379-387. 270, 219-227. 9. Lipton, S. A. & Rosenberg, P. A. (1994) N. Engl. J. Med. 330, 39. Deutsch, D. G. & Chin, S. (1993) Biochem. Pharmacol. 46, 613-622. 791-796. 10. Koh, J. Y., Goldberg, M. P., Hartley, D. M. & Choi, D. W. (1990) 40. Nadler, V., Mechoulam, R. & Sokolovsky, M. (1993) Neurosci. J. Neurosci. 10, 693-705. Lett. 162, 43-45. 11. Sheardown, M. J., Nielsen, E. O., Hansen, A. J., Jacobsen, P. & 41. Puttfarcken, P. S., Getz, R. L. & Coyle, J. T. (1993) Brain Res. Hon6re, T. (1990) Science 247, 571-574. 624, 223-232. 12. Schmid, H. H. O., Schmid, P. C. & Natarajan, V. (1990) Prog. 42. Kato, K., Puttfarcken, P. S., Lyons, W. E. & Coyle, J. T. (1991) J. Lipid Res. 29, 1-43. Pharmacol. Exp. Ther. 256, 402-411. 13. Natarajan, V., Schmid, P. C. & Schmid, H. H. 0. (1986) Biochim. 43. Gallo, V., Kingsbury, A., Balazs, R. & J0rgensen, O. S. (1987) J. Biophys. Acta 878, 32-41. Neurosci. 7, 2203-2213. 14. Epps, D. E., Natarajan, V., Schmid, P. C. & Schmid, H. H. O. 44. Balazs, R., Jorgensen, O. S. & Hack, N. (1988) Neuroscience 27, (1980) Biochim. Biophys. Acta 618, 420-430. 437-451. 15. Di Marzo, V., Fontana, A., Cadas, H., Schinelli, S., Cimino, G., 45. Rosenthal, H. E. (1967) Anal. Biochem. 20, 525-532. Schwartz, J.-C. & Piomelli, D. (1994) Nature (London) 372, 46. Hillard, C. J., Edgemond, W. S. & Campbell, W. B. (1995) J. 686-691. Neurochem. 64, 677-683. 16. Hansen, H. A., Lauritzen, L., Strand, A. M., Moesgaard, B. & 47. Howlett, A. C. (1995) Annu. Rev. Pharmacol. Toxicol. 35, 607- Frandsen, A. (1995) Biochim. Biophys. Acta 1258, 303-308. 634. 17. Hollister, L. E. (1986) Pharmacol. Rev. 38, 1-20. 48. Lynn, A. B. & Herkenham, M. (1994) J. Pharmacol. Exp. Ther. 18. Munro, S., Thomas, K.IL. & Abu-Shaar, M. (1993) Nature 268, 1612-1623. (London) 365, 61-65. 49. Pacheco, M. A., Ward, S. J. & Childers, S. R. (1993) Brain Res. 19. Facci, L., Dal Toso, R., Romanello, S., Buriani, A., Skaper, S. D. 603, 102-110. & Leon, A. (1995) Proc. Natl. Acad. Sci. USA 92, 3376-3380. 50. Aronica, E., Dell'Albani, P., Condorelli, D.F., Nicoletti, F., 20. Matsuda, L. A., Lolait, S. J., Brownstein, M. J., Young, A. C. & Hack, N. & Balazs, R. (1993) Mol Pharmacol. 44, 981-989. Bonner, T. I. (1990) Nature (London) 346, 561-564. 51. Paschen, W., Schmitt, J. & Uto, A. (1995) Soc. Neurosci. Abstr. 21. Gerard, C. M., Mollereau, C., Vassart, G. & Parmentier, M. 21, 765. (1991) Biochem. J. 279, 129-134. 52. Shen, M., Piser, T. M. & Thayer, S. A. (1995) Soc. Neurosci. Abstr. 22. Devane, W. A., HanuS, L., Breuer, A., Pertwee, R. G., Stevenson, 21, 1569. L. A., Griffin, G., Gibson, D., Mandelbaum, A., Etinger, A. & 53. Lipton, S. A. (1993) Trends Neurosci. 16, 527-532. Mechoulam, R. (1992) Science 258, 1946-1949. 54. Choi, D. W. (1990) Cerebrovasc. Brain Metab. Rev. 2, 105-147. 23. Felder, C. C., Briley, E. M., Axelrod, J., Simpson, J. T., Mackie, 55. Mercanti, D., Galli, C., Liguori, M., Ciotti, M. T., Gulla, P. & K. & Devane, W.A. (1993) Proc. Natl. Acad. Sci. USA 90, Calissano, P. (1992) Eur. J. Neurosci. 4, 733-744. 7656-7660. 56. Aloe, L., Leon, A. & Levi-Montalcini, R. (1993) Agents Actions 24. Skaper, S. D., Facci, L., Milani, D., Leon, A. & Toffano, G. 39, C145-C147. (1990) in Methods in Neurosciences, ed. Conn, P. M. (Academic, 57. Mazzari, S., Canella, R., Petrelli, L., Marcolongo, G. & Leon, A. San Diego), Vol. 2, pp. 17-33. (1996) Eur. J. Pharmacol., in press. Downloaded by guest on September 30, 2021